L'objectif de cette thèse est de mettre en évidence l'influence de la microstructure initiale sur le glissement intergranulaire lors du fluage à haute température d'un superalliage polycristallin à base de nickel. Dans ce but, plusieurs microstructures sont obtenues à partir de la microstructure de référence de l'alliage NR6, par application de traitements thermiques spécifiques. L'influence des paramètres microstructuraux sur les déformations locales est ensuite étudiée à l'aide d'une technique de microextensométrie couplée à une analyse par diffraction des électrons rétrodiffusés. Il est ainsi possible de relier microstructure, déformations locales et comportement macroscopique en fluage. Pour la microstructure de référence de l'alliage NR6, la déformation opère principalement par cisaillement des phases γ et γ'. Ce mécanisme est favorable au glissement intergranulaire. L'absence de précipités tertiaires de phase γ' favorise le contournement des précipités secondaires par les dislocations. Ceci permet de réduire le glissement intergranulaire mais est également néfaste pour la résistance à la déformation de l'alliage. La présence de joints de grains dentelés augmente la résistance au glissement intergranulaire mais diminue la résistance à la déformation intragranulaire en favorisant le contournement des précipités. Ainsi la résistance globale à la déformation n'est pas affectée. Enfin, l'augmentation de la taille de grains n'a d'influence ni sur les mécanismes de déformation mis en jeu ni sur l'amplitude du glissement. Cependant, la fraction moins élevée de joints de grains induit une diminution de la contribution du glissement intergranulaire à la déformation globale. / The aim of this study is to highlight the influence of initial microstructure on grain boundary sliding during high-temperature creep of a polycrystalline nickel-based superalloy. To reach this goal, several microstructures are produced from the reference microstructure of NR6 alloy by adequate heat treatments. The influence of microstructural parameters on local deformations is then studied thanks to a microextensometry technique coupled with an electron back-scattered diffraction analysis. It thereby enables linking microstructure, local deformations and macroscopic creep behaviour. In the case of NR6 alloy reference microstructure, deformation occurs mainly by γ and γ' phases cutting by dislocations. This mechanism is grain boundary sliding-favourable. The absence of tertiary γ' phase precipitates promotes secondary precipitates bypassing by dislocations. This results in a reduction of grain boundary sliding but is also harmful to the alloy creep resistance. Grain boundary serration improves grain boundary sliding resistance but diminishes intragranular deformation resistance by favouring precipitate bypassing. Then global deformation resistance is not changed. Finally, grain size increase has influence neither on activated deformation mechanisms nor on sliding amplitude. However, the decrease of grain boundary fraction leads to a reduction of grain boundary sliding contribution to overall strain.

Este estudo teve como objetivo principal desenvolver uma técnica para síntese de um nanocompósito de Y-TZP/MWCNT-COOH (Zircônia estabilizada por 3 mol% de ítria reforçada por nanotubos de carbono funcionalizado em -COOH) com propriedades mecânicas e ópticas que permitam a sua futura utilização como infraestrutura de próteses fixas dentárias e pilares protéticos para implantes. Assim, foram avaliados a microestrutura, resistência à flexão, tenacidade à fratura, limite de fadiga e propriedades ópticas do nanocompósito e comparada àquelas medidas para Y-TZP convencional (controle). O material Y-TZP/MWCNT-COOH foi desenvolvido pelo processo de co-precipitação de hidróxidos mistos associado ao tratamento hidrotérmico/solvotérmico e prensagem uniaxial em formato de blocos para sistemas CAD/CAM. O pó de MWCNT-COOH foi caracterizado por meio de MEV-FEG, TEM, TGA, DRX e FRX previamente a sua utilização para desenvolvimento do nanocompósito. Espécimes foram obtidos a partir do material Y-TZP/MWCNT-COOH para caracterização por meio de DRX, MEV-FEG e TEM, e comparação de suas propriedades estruturais (densidade e contração), ópticas, resistência à flexão, tenacidade à fratura e limite de fadiga com a Y-TZP convencional. O MWCNT-COOH apresentou-se em feixes de nanotubos de carbono recobertos por sílica tendo comprimento médio de 5,10 ± 1,34 ?m, com 90% dos comprimentos medidos (D90) estando abaixo de 6,9 ?m. Foi verificado a não possibilidade da utilização de líquidos orgânicos em nenhum passo da fabricação dos compósito Y-TZP/MWCNT-COOH por levar ao escurecimento do compósito, inviabilizando sua futura aplicação clínica. O tratamento hidrotérmico sem uso de líquidos orgânicos mostrou-se eficaz em proporcionar o revestimento do nanotubo de carbono por partículas de óxido de zircônio e ítrio. Entretanto, ocasionou a formação de aglomerados e partículas de Y-TZP com tamanho maiores que 5 ?m. Uma densidade relativa de 97,4% foi alcançada para o compósito experimental de Y-TZP contendo MWCNT-COOH, tendo uma razão de contraste de 0.9929 ± 0.0012 e um valor de diferença de cor da Y-TZP convencional de 6,1 ± 3,1 ( ?E). As propriedades mecânicas da Y-TZP/MWCNT-COOH, dureza Vickers (10,14 ± 1,27 GPa; p=0,25) e tenacidade à fratura (4,98 ± 0,30 MPa.m1/2; p=0,39), não apresentaram diferença significativa da Y-TZP convencional (dureza: 8,87 ± 0,89; tenacidade à fratura: 4,98 ± 0,30 MPa.m1/2). Entretanto, para a resistência à flexão (p=0,003) e limite de fadiga cíclica (LFC) foram obtidos valores inferiores para o material experimental Y-TZP/MWCNT-COOH (resistência à flexão: 299,4 ± 30,5 MPa; LFC: 179,4 ± 22,5 MPa) quando comparado à Y-TZP controle (resistência à flexão: 623,7 ± 108,8 MPa; LFC: 439,0 ± 56,4 MPa). Com base nos resultados apresentados, é possível concluir que a síntese de um nanocompósito de Y-TZP/MWCNT-COOH com propriedades ópticas adequadas para aplicação na odontologia restauradora foi possível por meio dos métodos descritos, entretanto algumas adequações nos métodos de síntese e processamento para criação do nanocompósito devem ser realizadas para se evitar a acentuada diminuição de importantes propriedades mecânicas do material. / This study aim was to develop a technique for synthetize nanocomposite of Y-TZP/MWCNT-COOH (3 mol% Yttria-Stabilized Tetragonal Zirconia reinforced with COOH functionalized carbon nanotubes) with mechanical and optical properties that allow their future use as fixed dental prosthesis infrastructure and implant abutments. Thus, the following properties of the nanocomposite were investigated and compared to those measured for conventional Y-TZP (control): microstructure, flexural strength, fracture toughness, fatigue limit and optical properties. Y-TZP/MWCNT-COOH material was developed by the co-precipitation of mixed hydroxides associated with the hydrothermal/solvothermal treatment and uniaxial pressing to form blocks for CAD/CAM systems. The MWCNT-COOH powder was characterized by SEM-FEG, TEM, TGA, XRD and XRF prior to its use for the development of nanocomposite. Specimens were obtained from the Y-TZP/MWCNT-COOH material and characterized by XRD, SEM-FEG and TEM. After characterization, the material had their structural properties (density and contraction), optical, flexural strength, fracture toughness and fatigue limit compared to a conventional Y-TZP. The MWCNT-COOH material was observed to be a bundle formation of carbon nanotube covered with silica with an average length of 5.10 ± 1.34 ?m, with 90% of the measured lengths (D90) being below 6.9 ?m. It has been found to be not possible to use organic liquids on any step of the Y-TZP/MWCNT-COOH manufacturing process due to darkening of the composite, making it unfeasible to future clinical application. The hydrothermal treatment without the use of organic liquids was effective in providing the carbon nanotube coating by zirconium and yttrium oxide particles. However, this treatment led to the formation of agglomerates and particles of Y-TZP with larger than 5 ?m. A relative density of 97.4% was achieved for the Y-TZP/MWCNT-COOH composite, having a contrast ratio of 0.9929 ± 0.0012, and a color difference value from the conventional Y-TZP of 6.1 ± 3.1 (?E). The mechanical properties of Y-TZP/MWCNT-COOH, Vickers hardness (10.14 ± 1.27 GPa; p = 0.25) and fracture toughness (4.98 ± 0.30 MPa.m1/2; p = 0.39), showed no significant difference from the conventional Y-TZP (hardness: 8.87 ± 0.89; fracture toughness: 4.98 ± 0.30 MPa.m1/2). However, flexural strength (p = 0.003) and cyclic fatigue limit (CFL) showed lower values for the experimental material Y-TZP/MWCNT-COOH (flexural strength: 299.4 ± 30.5 MPa; CFL: 179.4 ± 22.5 MPa) compared to Y-TZP control (flexural strength: 623.7 ± 108.8 MPa; CFL: 439.0 ± 56.4 MPa). Based on the results presented, it could be conclude that the synthesis of a nanocomposite of Y-TZP/MWCNT-COOH with optical properties suitable for application in restorative dentistry was made possible by the described methods, however some adjustments in synthesis and processing methods for the nanocomposite creation should be taken; to avoid the sharp decrease of important mechanical properties of the material.

Les méthodes de calcul en fatigue à grande durée de vie sont en cours de développement depuis des décennies et sont utilisées par les ingénieurs pour dimensionner les structures. Généralement, ces méthodes se basent sur la mise en équations de quantités mécaniques calculées à l'échelle macroscopique ou mésoscopique. Les critères de fatigue multiaxiale reposent généralement sur des hypothèses de changement d'échelle dont l'objectif est d'accéder à l'état de contraintes ou de déformations à l'échelle du grain. Dans les approches de type plan critique (Dang Van, Papadopoulos, Morel), l'amorçage d'une fissure de fatigue est considéré comme piloté par une quantité mécanique liée à une orientation matérielle particulière (plan critique). Si ces phénomènes sont bien établis dans le cas des chargements uniaxiaux, la nature des mécanismes liés à l'activation des systèmes de glissement, à la multiplicité du glissement et aux différents sites préférentiels d'amorçage de fissures sous chargements complexes reste peu connue.Afin de mieux comprendre les mécanismes d'endommagement en fatigue multiaxiale, les techniques d'analyse et de caractérisation de l'activité plastique (activation des systèmes de glissements, bandes de glissement persistantes) et d'observation de l'endommagement par fatigue ont été mises en place en se basant principalement sur des observations MEB et analyses EBSD. Ces investigations ont permis de mettre en lumière les effets des chargements non proportionnels sur la multiplicité du glissement sur du cuivre pur OFHC. L'étude statistique des sites préférentiels d'amorçage de fissures montre que les grains à glissement multiple présentent une forte probabilité d'amorçage de fissures, surtout sous les chargements non proportionnels. Nous avons également mis en évidence le rôle des joints de grains et des joints de macle sur le développement de la plasticité à l'échelle de la microstructure. Les résultats expérimentaux sont confrontés à ceux du calcul éléments finis (EF) en plasticité polycristalline sur des microstructures synthétiques 3D semi-périodiques. L'application du critère de Dang Van à l'échelle mésoscopique (le grain) montre une forte variabilité de la contrainte hydrostatique et du cisaillement. Cette variabilité est plus importante pour un modèle de comportement cristallin élastique anisotrope. Le rôle de la plasticité cristalline se révèle secondaire. Ces analyses permettent de remettre en perspective les hypothèses usuelles de changement d'échelle utilisées en fatigue multiaxiale. Enfin, une méthode basée sur la statistique des valeurs extrêmes est proposée pour le dépouillement des calculs EF sur agrégats. Cette analyse a été appliquée sur la contrainte équivalente associée au critère de fatigue de Dang Van pour les calculs d'agrégats polycristallins avec différentes morphologies et orientations des grains. Les effets de la surface libre, du type de chargement et du modèle de comportement mécanique des grains ont été analysés. Les résultats offrent des perspectives intéressantes sur la modélisation de l'amorçage des fissures en fatigue multiaxiale des matériaux et des structures avec une prise en compte de la microstructure. / The development of high cycle fatigue (HCF) strength assessment methods has now been running for more than a century, leading to relatively efficient methods for engineers. Generally, these methods are based on mechanical quantities calculated at macroscopic or mesoscopic scales and validated by the model's ability to accurately reproduce experimental results. Multiaxial fatigue strength criteria are usually based on scaling transition assumptions aiming at capturing the stress or strain state in the grain. In the case of critical plane based criteria (Dang Van, Papadopoulos, Morel), fatigue crack initiation is supposed to be controlled by a mechanical quantity linked to a particular orientation (critical plane). If fatigue crack initiation phenomena are well established in the case of uniaxial loadings, the nature of the mechanisms involved in the activation of slip systems, multiple slip and preferential sites of rack initiation under complex loadings remains little known.To better understand the mechanisms of multiaxial fatigue crack initiation, analysis and characterization of the plastic activity (e.g. activation of slip systems, persistent slip bands) and observations of fatigue damage have been carried out on pure OFHC copper, using SEM and EBSD analyses. These investigations enabled to highlight the effects of non-proportional multiaxial loadings through the induced multiplicity of slip. The statistical study of preferential crack initiation sites shows that grains with multiple slip have a high probability of crack initiation, especially under non-proportional loading. We also highlighted the role of grain boundaries and twin boundaries on the development of plasticity across the microstructure. The experimental results were compared with those of finite element crystal plasticity computations on synthetic 3D semi-periodic microstructures. The application of the Dang Van criterion at the mesoscopic (grain) scale showed a strong variability of the hydrostatic stress and the shear stress. This variability was greater for anisotropic elastic behavior, while the role of crystal plasticity seemed to be secondary. These analyses allowed putting into perspective the usual assumptions of scaling transition rules used in multiaxial fatigue. Finally, a method based on the extreme values statistics ​​was proposed and applied to the equivalent stress associated to the Dang Van fatigue criterion for polycrystalline aggregate computations with different morphologies and grains orientations. The effects of the microstructure, free surface, loading types and mechanical behavior were analyzed. The results offered interesting insights into the multiaxial fatigue modeling of metals and structures taking into account the microstructure.

L'objectif de ces travaux de thèse est d'étudier l'influence de la microstructure et de défauts géométriques sur le comportement en fatigue à grand nombre de cycles (FGNC) d'un acier inoxydable austénitique 316L. La méthodologie proposée s'appuie sur des simulations par éléments finis (EF) d'agrégats polycristallins qui permettent de décrire les champs mécaniques à l'échelle des mécanismes impliqués dans les processus d'amorçage de fissures de fatigue.Une étude numérique préliminaire, s'appuyant sur des données expérimentales issues de la littérature, est conduite sur un cuivre électrolytique à l'aide de simulations numériques d'agrégats polycristallins en 2D. L'effet du trajet de chargement et de défauts artificiels de taille proche ou légèrement supérieure à celle de la microstructure sur les réponses mécaniques mésoscopiques sont analysés. Les capacités de prédiction de quelques critères de fatigue, s'appuyant sur des quantités mécaniques mésoscopiques, sont évaluées. Il est mis en évidence que les limites de fatigue macroscopiques prédites par un critère de fatigue probabiliste sont en accord avec les tendances expérimentales observées en fatigue multiaxiale et en présence de défauts.Une campagne expérimentale a été menée sur un acier austénitique 316L. Des essais de fatigue oligocyclique sont conduits afin de caractériser le comportement élasto-plastique du matériau. Des essais de FGNC, utilisant des éprouvettes avec et sans défaut de surface (défaut artificiel hémisphérique) ont été effectués pour estimer les limites de fatigue dans différentes conditions de sollicitation (traction, torsion, traction et torsion combinée, traction biaxiale) et pour plusieurs rayons de défaut. Dans le but de compléter la caractérisation du matériau, la microstructure est étudiée à l'aide d'analyses EBSD et la texture cristallographique est mesurée par diffraction des rayons X. Ces résultats expérimentaux sont utilisés pour reproduire, avec des simulations EF, les essais de FGNC sur des microstructures 2D et 3D représentatives de l'acier austénitique. L'hétérogénéité de quantités mécaniques mésoscopiques pertinentes en fatigue est discutée avec une attention particulière sur l'effet des défauts. L'approche probabiliste est appliquée aux résultats des modèles EF pour quantifier l'effet de la taille du défaut, pour différents trajets de chargement. La pertinence, vis-à-vis des observations expérimentales, des distributions de la limite de fatigue prédites est évaluée. / The aim of this study is to analyze the influence of both the microstructure and defects on the high cycle fatigue (HCF) behaviour of a 316L austenitic stainless steel thanks to finite element (FE) simulations of polycrystalline aggregates.%The scatter encountered in the HCF behavior of metallic materials is often explained by the anisotropic elasto-plastic behavior of individual grains leading to a highly heterogeneous distribution of plastic slip.Since fatigue crack initiation is a local phenomenon, intimately related to the plastic activity at the crystal scale, it seems relevant to rely on this kind of modeling to evaluate the mechanical quantities.A preliminary numerical study, based on experimental data drawn from the litterature, was conducted on an electrolytic copper using simulations of 2D polycrystalline aggregates. The effect of the loading path and small artificial defects on the mesoscopic mechanical responses have been analyzed separately. Moreover, the predictive capabilities of some fatigue criteria, relying on the mesoscopic mechanical responses, has been evaluated. It was shown that the macroscopic fatigue limits predicted by a probabilistic fatigue criterion are in accordance with the experimental trends observed in multiaxial fatigue or in the presence of small defects.An experimental campaign is undertaken on an austenitic steel 316L. Low cycle fatigue tests are conducted in order to characterize the elasto-plastic behavior of the material. Load-controled HCF tests, using both smooth specimens and specimens containing an artificial hemispherical surface defect, are carried out to estimate the fatigue limits under various loading conditions (tension, torsion, combined tension and torsion, biaxial tension) and several defect radii. To complete the characterization of the material, the microstructure is studied thanks to EBSD analyzes and the cristallographic texture is measured by X-ray diffraction. These experimental data are used to reproduce, with FE simulations, the HCF tests on 2D and 3D microstructures representative of the austenitic steel. The heterogeneity of the mesoscopic mechanical quantities relevant in fatigue are discussed in relation to the modeling. The results from the FE models are then used along with the probabilistic mesomechanics approach to quantify the defect size effect for several loading paths. The relevance, with respect to the experimental observations, of the predicted fatigue strength distributions is assessed.

The objective of this study is analysing the characteristics of five different solar cell technologies regarding their efficiency, fill factor, cost and environmental impacts and comparing their improvement records over years considering their efficiency. The five solar cell technologies of interest are amorphous silicon, monocrystalline silicon, polycrystalline silicon, cupper indium gallium selenide thin film and cadmium telluride thin film. The structure and manufacturing process of each of cell technologies were discussed. The study was conducted by the aid of available scientific reports regarding the electrical characteristics of different solar cell technologies. The extracted information regarding efficiency rate and fill factor was analysed using graphs and significant findings are discussed. The five technologies are also compared regarding their cost and ease of fabrication and their impacts on environment and recycling challenges. The result of this study is suggesting the most promising technology that may be the optimal option for further investment and research.

Transition Metal Silicides (TMS) are well known refractory materials because of their high thermal and structural stability at elevated temperature. In addition TMS materials are known for their moderate thermoelectric applications at high temperature since they exhibit superior semiconducting behavior. But TMS materials have relatively higher thermal conductivity which limits their applications in the field of thermoelectrics. So it is important to reduce their thermal conductivity to enhance conversion efficiency. In this regard, the work is performed to reduce the thermal conductivity of selected silicides such as CrSi2, MnSi2, and β-FeSi2 through alloys scattering and nano-structuring by mechanical alloying. A brief introduction about basic principles of thermoelectricity and related parameters are described in the chapter 1. Thermoelectric material’s figure of merit (zT) depends on the ratio of carrier charge transport and thermal energy transport. The conversion efficiency can be significantly enhanced by increasing the zT value. This chapter discusses the methods to increase the zT and list out some of the state-of-art of thermoelectric materials which possesses high zT value. Chapter 2 covers the preparation of selected silicides, such as CrSi2, MnSi2 and β-FeSi2, and the characterization techniques used to define the thermoelectric performance. In this chapter the suitability and the performance of transition metal silicides for high temperature thermoelectric application are discussed. In summary, the objective of the thesis has been framed. Chapter 3 deals with thermoelectric properties of pure and Mn, Al doped chromium disilicide (CrSi2). This chapter has been divided into three parts and discussed the effect of composition variation (CrSi1.90-2.10), point defects (by introducing Al at Si site), and mass-fluctuation scattering (by co-substitution of Mn and Al) on thermoelectric properties of polycrystalline CrSi2 in the temperature range of 300K-800K. In the first part, it is observed that CrSi2 has a homogeneity range of CrSi1.95-CrSi2.02. The secondary phases evolve above and below this homogeneity range. These secondary phases significantly scatter phonons and reduce the thermal conductivity. In the second part, Al has been introduced at Si site in CrSi2 and creates the point defects which is also scatter the short wavelength phonons and lead to low thermal conductivity. The third part explores the influence of co-substitution of Mn at Cr site and Al at Si site on lattice thermal conductivity. Here, substitution of Al creates point defects and addition of Mn leads to mass fluctuation scattering. These combined effects result in huge reduction in lattice thermal conductivity and thereby enhanced the zT. Chapter 4 deals with efforts of nano-structuring the CrSi2 through Mechanical Alloying (MA) using SS (stainless steel) and WC (Tungsten Carbide) milling media. The effects of two milling media on crystallite size reduction are discussed. It is seen that as milling time increases the rate of crystallite size reduction also increases. The X-ray diffraction studies of hot pressed pellets show the formation of secondary metallic phase like Cr1-xFexSi from SS milled samples and CrSi from WC milled samples. It indicates that CrSi2 gains metallic Fe atoms during mechanical alloying and the secondary phases are formed. As milling time increases it is observed that weight loss from the milling balls also increases. The Fe content coming from SS ball forms a solid solution with CrSi phase. The transport properties like resistivity, Seebeck coefficient and thermal conductivity were measured for milled samples from 300K-800K. It is observed that formation of the secondary metallic phase reduces resistivity and Seebeck coefficient of overall ceramics. Very large reduction in thermal conductivity was found for samples which were 15hrs-WC-milled (7.4 W/m.K at 375K) due to increased phonon scattering by grain boundaries. The 15hrs-SS-milled samples show thermal conductivity ~10 W/m.K which is considerably low as compared to the as-cast CrSi2 (13.5 W/m.K). This chapter explores the structural studies and mechano-chemical decomposition of CrSi2. In addition, the influences of mechanical milling media and micron size secondary phase on transport properties of CrSi2 are also discussed. Chapter 5 deals with the influence of microstructures of MnSi2 densified by hot uni-axial pressing (HP) and spark plasma sintering (SPS) on thermoelectric properties. The effects of these densification processes on arresting the grain growth during sintering are explored. The powder X-ray diffraction studies show higher manganese silicide (HMS) with secondary Si phase. The SEM and EPMA results confirmed the presence of Si phase. The TEM micrographs are shown the particle size distribution of HMS to be <200nm with fine precipitates of Si, of 5-10nm size, in the HMS matrix. The ball milled samples of MnSi2 showed increase in resistivity and Seebeck coefficient with large reduction in total thermal conductivity as compared to that seen in as-cast sample. The SPS densified samples show lower thermal conductivity, with reduction by about 52%, as compared to HP sample’s (45%) reduction for same conditions. An enhancement in zT by 73% could be achieved for the SPS densified for 2 min at 1060˚C. Chapter 6 examines (i) the decomposition of α–FeSi2, generally known as α-Fe2Si5, (eutectoid reaction) into β-FeSi2 with Si dispersoids (ii) formation of β-FeSi2 from ε-FeSi and α-Fe2Si5 (peritectoid reaction). This is accompanied by a discussion of the microstructural effect on thermoelectric properties. Prolonged annealing of peritectoid composition decomposes the α– FeSi2 phase, replaces the ε–FeSi phase, and forms pure β-FeSi2 whereas eutectoid composition of α–FeSi2 decomposes into lamellar structure of β-FeSi2 and Si dispersions. The aging heat treatment carried out for composition prepared from eutectoid reaction at various temperatures (600°C, 700°C, 800°C and 850°C for duration of 100hrs, 10hrs, 4hrs and 10hrs, respectively) below the equilibrium eutectoid temperature were found to have fine and homogenous dispersions of Si particles. The XRD and SEM studies confirmed the presence of a secondary Si phase on the matrix of β-FeSi2 for the heat treated eutectoid composition. The excess Si phase in β-FeSi2 increases the resistivity and Seebeck coefficient by the reducing carrier concentration of system as compared to those that of pure β-FeSi2, which is prepared from peritectoid composition. The samples heat treated at 600°C showed relatively low thermal conductivity as compared to that of β-FeSi2. This chapter gives a route map for reducing the thermal conductivity by micro structural engineering through Si dispersions on β-FeSi2. In addition, this comparison of two the decomposition processes and its influence on the microstructure and thermoelectric properties is made. Chapter 7 summarizes the key conclusions of the work performed in this thesis. The work reported in this thesis has been carried out by the candidate as a part of Ph.D training programme. He hopes that this would constitute a worthwhile contribution to the field of thermoelectrics for understanding the (i) effect of alloy scattering, (ii) mass fluctuation scattering, (iii) and nano-structuring of transition metal silicides for high temperature thermoelectric materials.

The densification mechanisms of polycrystalline α-alumina by spark plasma sintering are highly contradictory, with different research groups suggesting diffusion to dislocation controlled mechanisms to be rate controlling. The specific objective of this work was to investigate densification mechanisms of α-alumina during the intermediate and final stages of sintering by SPS, analyze the microstructural development and establish sintering trajectories. In addition, zirconia and yttria were added in different weight percentages to study the effect of solute concentration on the densification kinetics of spark plasma sintered alumina. The present work adopts a different approach from the classical method adopted previously to analyze the sintering kinetics and densification mechanisms of alumina in SPS, although existing models for hot pressing were adopted for the basic analysis. The densification behavior was investigated in the temperature range 1223-1573 K under applied stresses of 25, 50 and 100 MPa and grain sizes between 100 and 250 nm. The SEM micrographs reveal equiaxed grains with no abnormal grain growth in the dense samples. The ‘master sintering curve’ shows grain size to be primarily dependent on density, irrespective of the applied stresses or temperature. The stress exponent of 1 along with an inverse grain size exponent of 3 and activation energy of 320-550 kJ mol-1 suggests Al3+ grain boundary diffusion as the rate controlling densification mechanism in alumina. The densification rates are marginally slower in compositions with 0.1% Y2O3 and ZrO2 content possibly due to the smaller grain sizes used in this study which leads to faster rates compared to earlier reports. However, higher Y2O3 and ZrO2 content led to decrease in densification rate by more than an order of magnitude possibly due to presence of a second phase which increases the effective path length for diffusion, thereby reducing the densification rates. Presence of Y2O3 and ZrO2 in the compositions with 0.1% Y2O3 and ZrO2 were confirmed by TEM studies. The Y3Al5O12 (YAG) phase developed between 1223 and 1273 K and suppressed densification and grain growth in alumina. In spite of higher temperatures required for alumina-YAG and alumina-zirconia composites to attain density ~99%, the alumina grain size in the composites was smaller than that in pure alumina due to the Zener drag effect. The stress exponents obtained for Y2O3 and ZrO2 composites at both the concentrations yield a value of n~ 2, which indicates a change in densification mechanism from pure alumina. The higher stress dependence of these composites could be due to presence of solute and second phase formation, both of which retard densification rates. The inverse grain size exponents obtained are between 1 and 2; both stress exponent and grain size exponent values suggest an interface reaction controlled diffusion mechanism occurring in these composites, independent of the Y2O3 and ZrO2 content. Higher activation energies are obtained with the Y2O3 and ZrO2 composites of higher content, respectively, due to presence of second phase particles at grain boundaries. The presence of solutes at grain boundaries hinders grain boundary diffusion of alumina, leading to interface reaction controlled process; this is confirmed by superimposing standard aluminum grain boundary and lattice diffusion data on to stress-densification rate data obtained in this work. A comparison of stress exponents using current experimental data adopting the present and the classical approaches show a wide difference in their values indicating a change in the rate controlling diffusion path, necessitating a review of the assumptions made on the basic equations used in previous SPS studies.

La thèse s'inscrit dans le cadre du projet GENOME « Gestion Optimisée de l'Energie » dont l'enjeu majeur est la conception d'un avion plus électrique. L'augmentation de l'efficacité énergétique et de la compacité des systèmes électriques de ces avions entraîne de fortes sollicitations en température. Ces sollicitations sont liées à la compacité des systèmes (réduction de masse et de volume) ainsi qu'à leur localisation par rapport aux sources chaudes (réacteur d'avion par exemple). De ce fait, les matériaux magnétiques des nouveaux convertisseurs électriques doivent pouvoir fonctionner sous des conditions de hautes températures, supérieures à 200°C. Il s'agit du polycristallin FeCoV dédié à la fabrication des tôles du stator et du rotor des génératrices rapides (situées à proximité des réacteurs) et le nanocristallin FeCuNuSiB dédié à la conception des inductances et transformateurs des convertisseurs statiques. Ce manuscrit s'intéresse à l'étude du vieillissement thermique de ces deux familles de matériaux magnétiques. Ces matériaux, fournis par la société APERAM, se déclinent sous plusieurs nuances et finitions. L'étude du vieillissement consiste en l'application de plusieurs essais de vieillissement continus sous différentes températures (jusqu'à 300 °C pour les FeCoV et 240 °C pour les nanocristallins). Plusieurs grandeurs macroscopiques magnétiques, électriques et mécaniques (pour les FeCoV) sont mesurées à chaque intervalle de vieillissement. Grâce à ces mesures macroscopiques et à des mesures complémentaires effectuées à l'échelle microscopique, des analyses sont faites et des hypothèses sont proposées afin d'expliquer les mécanismes de vieillissement de ces deux familles de matériaux et dans le but de proposer des modèles phénoménologiques fiables / The thesis takes part of the project GENOME “Gestion Optimisée de l’Energie” whose major issue is the design of the more electrical aircraft. The increase in the energy efficiency and the compactness of the electrical systems of these aircrafts lead to high temperature stresses. These thermal stresses are related to the compactness of the systems (reduction of mass and volume) as well as their location with respect to the hot sources (aircraft engine for example). Thus, the magnetic materials of the new electrical converters must be able to operate under conditions of high temperatures, above 200 °C. Typically, the FeCoV polycrystalline materials are dedicated to the fabrication of the stator and rotor sheets of the fast generators (located near the aircraft engine) and the FeCuNbSiB nanocrystalline materials are dedicated to the design of inductors and transformers of the static converters.This manuscript concerns the thermal ageing study of these two magnetic material families. These materials, supplied by the company APERAM, are available in several shades. The ageing study consists on applying several continuous ageing treatments at different temperatures (up to 300 °C for FeCoV and 240 °C for FeCuNbSiB). At each ageing step, several macroscopic properties namely: magnetic, electrical and mechanical (for the FeCoV materials) properties are measured. Using these macroscopic properties and complementary measurements carried out on a microscopic scale, analyses are made and hypotheses are proposed in order to explain the ageing mechanisms of these magnetic material families. The understanding of the magnetic ageing mechanisms is necessary towards establishing of phenomenological ageing models

An understanding of crack tip fields and fracture mechanisms in single crystals can help in developing better polycrystalline alloys and manufacturing processes. To this end, the effects of loading rate, material inertia and strain rate sensitivity on crack tip fields and their influence on fracture mechanisms in FCC single crystals are examined in this work by performing finite element analysis. It is shown that, in the absence of inertial effects, high loading rates elevate the stresses ahead of a crack tip and decrease the plastic strains in rate dependent single crystals. Also, it is found that the quasi-static near-tip stress field can be adequately characterized by the energy release rate J and a constraint parameter Q. Similar two-parameter characterization is possible even under dynamic loading. It is observed that if a suitable reference solution is used, the role of inertia manifests as a loss of constraint with increasing loading rate irrespective of strain rate sensitivity and lattice orientation. Thus, at very high loading rates, inertial effects oppose the role of rate sensitivity and cause a decrease in stresses near the tip. The relative influence of these two factors depends on rate sensitivity index. For a mildly rate dependent single crystal, the predicted cleavage fracture toughness remains constant up to a certain loading rate and thereafter increases sharply. On the other hand, for a strongly rate dependent single crystal, fracture toughness drops initially up to a certain loading rate beyond which it increases marginally. The loss of crack tip constraint is found to retard the ductile fracture mechanisms of void growth and coalescence. However, this is dependent on lattice orientation. In-situ experimental observation of void growth near a notch tip also shows strong orientation dependence. In addition, 3D finite element results indicate though-thickness dependence of equivalent plastic slip and hydrostatic stress leading to variations in void growth along the thickness direction of the specimens. The predicted load-displacement curves, lattice rotation, slip traces and void growth using finite element analysis are found to be in good agreement with the experimental observations. Thus, the present study has provided an understanding of the role of several factors such as constraint level, rate sensitivity, material inertia, lattice orientation and 3D effects on the mechanics of fracture of ductile single crystals.

The macroscopic behavior of complex heterogeneous materials is strongly governed by the interactions between their elementary constituents within their microstructure. Beside experimental efforts characterizing the behaviors of such materials, there is growing interest, in view of the increasing computational power available, in building models representing their microstructural systems integrating the elementary behaviors of their constituents and their geometrical organization. While a large number of contributions on this aspect focus on the investigation of advanced physics in material parameter studies using rather simple geometries to represent the spatial organization of heterogeneities, few are dedicated to the exploration of the role of microstructural geometries by means of morphological parameter studies.<p>The critical ingredients of this second type of investigation are (I) the generation of sets of representative volume elements ( RVE ) describing the geometry of microstructures with a satisfying control on the morphology relevant to the material of interest and (II) the discretization of governing equations of a model representing the investigated physics on those RVEs domains. One possible reason for the under-representation of morphologically detailed RVEs in the related literature may be related to several issues associated with the geometrical complexity of the microstructures of considered materials in both of these steps. Based on this hypothesis, this work is aimed at bringing contributions to advanced techniques for the generation and discretization of microstructures of complex heterogeneous materials, focusing on geometrical issues. In particular, a special emphasis is put on the consistent geometrical representation of RVEs across generation and discretization methodologies and the accommodation of a quantitative control on specific morphological features characterizing the microstructures of the covered materials.<p>While several promising recent techniques are dedicated to the discretization of arbitrary complex geometries in numerical models, the literature on RVEs generation methodologies does not provide fully satisfying solutions for most of the cases. The general strategy in this work consisted in selecting a promising state-of-the-art discretization method and in designing improved RVE generation techniques with the concern of guaranteeing their seamless collaboration. The chosen discretization technique is a specific variation of the generalized / extended finite element method that accommodates the representation of arbitrary input geometries represented by level set functions. The RVE generation techniques were designed accordingly, using level set functions to define and manipulate the RVEs geometries. <p>The RVE methodologies developed are mostly morphologically motivated, incorporating governing parameters allowing the reproduction and the quantitative control of specific morphological features of the considered materials. These developments make an intensive use of distance fields and level set functions to handle the geometrical complexity of microstructures. Valuable improvements were brought to the RVE generation methodologies for several materials, namely granular and particle-based materials, coated and cemented geomaterials, polycrystalline materials, cellular materials and textile-based materials. RVEs produced using those developments have allowed extensive testing of the investigated discretization method, using complex microstructures in proof-of-concept studies involving the main ingredients of RVE-based morphological parameter studies of complex heterogeneous materials. In particular, the illustrated approach offers the possibility to address three crucial aspects of those kinds of studies: (I) to easily conduct simulations on a large number of RVEs covering a significant range of morphological variations for a material, (II) to use advanced constituent material behaviors and (III) to discretize large 3D RVEs. Based on those illustrations and the experience gained from their realization, the main strengths and limitations of the considered discretization methods were clearly identified. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished